Experimental and Numerical Study on Performance of Undowelled Joints In Concrete Pavements

Abstract

Joints are the major structural feature of concrete pavements, which play a major role in the effective and efficient performance of pavements. The load transfer across the transverse joints occurs mainly through dowel bars in jointed plain cement concrete pavements (JPCPs). However, in whitetopping and short-panelled concrete pavements, the joints are undowelled, and load transfer mainly occurs through aggregate interlocking. In India, there are no specific guidelines available for the design of short- panelled concrete pavements. In some of the projects, IRC SP:76-2015, which is the guidelines for the design and construction of whitetopping, was followed to design and construct short-panelled concrete pavements. The guideline IRC SP:76-2015 does not account for the load transfer across the joint as a critical factor for the design. It is assumed that the underlying pavement provides considerable support to the joints, which improves the load transfer across the joint. However, it was reported that the poor load transfer across the aggregate interlocked joints results in debonding with the underlying bituminous layers in the case of whitetopping. The poor performance of aggregate interlocking results in faulting, spalling and corner breaks in short-panelled concrete pavements. Since, the performance of the whitetopping and short-panelled concrete pavement depends on the performance of aggregate interlocked joints, the parameters influencing the performance of aggregate interlocked joints have to be evaluated at the design stage and at the time of service. In the present study, a new test methodology is proposed to characterise the shear transfer ability of aggregate interlocking in plain, micro and macro-fiber reinforced pavement quality concrete (PQC) cylindrical specimens by conducting the direct shear test in the laboratory. The influence of the nominal maximum aggregate size (NMAS), fiber dosages and groove depth (GD) on shear strength (τ), joint shear stiffness (K), and fracture energy in mode - II (GIIF) of aggregate interlocking in pavement quality concrete (PQC) mixes are studied under the static loading. A relationship between GIIF and K is determined for each of the PQC mixes. Also, a shear fatigue test is conducted at higher stress levels to evaluate the effect of NMAS, fiber dosages and GD on the performance of aggregate interlocking in PQC specimens at the grooved cross-section. Statistical analysis was carried out to understand the influence vii of GD and stress levels on the fatigue performance of plain, micro, and macro-fiber reinforced PQC mixes. A small-scale test setup was developed to simulate the aggregate interlocking of concrete pavements. The simulated test setup was used to evaluate the performance of aggregate interlocking in terms of load transfer efficiency (LTE). The beam specimens were utilised to carry out this test. The relationship between the relative movement of the beam (difference in deflection of loaded and unloaded side) and LTE is developed for each PQC mix and is compared with the field FWD results for validation. The 3D FE models are also developed in ANSYS for the two tests proposed in this study. The 3-D finite element model results are in good agreement with the experimental results for both proposed tests. In the present study, an improvement to the existing analytical model is also presented that can be used to compute the joint stiffness of white-topping pavements directly from the FWD deflection data. Further, ANN models have been developed and compared for the proposed and previously available analytical models in the literature. The joint stiffness calculated from the ANN model developed from the proposed analytical model is used as an input parameter in the FE model, and LTE is compared with the field studies. The proposed ANN model is simple, efficient and accurate enough to estimate the joint stiffness directly from FWD deflection data. From the experimental tests carried out in this research work, it is concluded that the proposed test methods can be used to evaluate undowelled joints in concrete pavements. Also, from the laboratory studies, it is concluded that larger NMAS (31.5 mm) and dosage of macro-fiber (0.75%) can substantially improve the aggregate interlocking in short panelled concrete pavements, and the use of fiber dosage up to 2.1 kg/m3 micro-fibers can substantially improve the undowelled/aggregate interlocked joint performance of whitetopping when compared to other dosages under study.